PolyPhen-2

PolyPhen-2 predicts the functional impact of amino acid substitutions on human protein stability and function by integrating structural analysis, evolutionary conservation, and machine-learning classification.

Key Features:

• Functional Annotation: Maps coding single-nucleotide polymorphisms (SNPs) to gene transcripts and annotates resulting amino acid substitutions.
• Sequence and Structural Analysis: Extracts protein sequence annotations and structural attributes relevant to amino acid changes.
• Conservation Profiles: Builds conservation profiles to assess evolutionary constraints on residues using vertebrate MultiZ alignments to the human genome.
• Machine Learning Classification: Employs a machine-learning classifier to estimate the probability that a missense mutation is damaging.
• High-Quality Alignments: Produces high-quality multiple protein sequence alignments for comparative analysis across species.
• Integration with Genome Resources: Integrates annotations from the UCSC Genome Browser and utilizes MultiZ vertebrate alignments with the human genome.
• High-Performance and Large-Scale Processing: Handles large datasets from next-generation sequencing projects and supports execution in high-performance computing environments.

Scientific Applications:

• Variant Effect Interpretation: Interpreting the functional consequences of missense variants and coding SNPs in human proteins.
• Disease Research: Investigating mutations identified in disease studies to inform understanding of disease mechanisms.
• Genomic-Scale Prioritization: Prioritizing candidate variants from next-generation sequencing and large-scale genomic studies.

Methodology:

Combines structural data, sequence annotation, conservation profiles derived from MultiZ vertebrate alignments and high-quality multiple sequence alignments, and a machine-learning classifier to estimate the probability that a missense mutation is damaging.